[H-DS08] Landslides and related phenomena

[HDS08-P07] Slope disaster detection and triggering factor analysis via integrated use of time-series optical and SAR satellite data in Kyushu Island, Japan

*Hiroki Mizuochi¹, Moe Matsuoka¹, Satoru Yamamoto¹, Kazuhiro Miyazaki¹, Tomoya Abe¹, Hideo Hoshizumi¹, Daisaku Kawabata¹, Koki Iwao¹, Yoshinori MIYACHI¹ (1.National Institute of Advanced Industrial Science and Technology)

Keywords:Slope disaster monitoring, long-term optical data, InSAR analysis, trend analysis

Recently, satellite remote sensing is an essential tool for slope disaster observation. Taking advantage of large data archive of optical and microwave (i.e., synthetic aperture radar: SAR) satellite images, we provided time-series analysis of sloping area in Kyushu Island, which is one of the high-risk areas of slope disaster in Japan. Specifically, anomaly detection algorithm using the open-free optical data (i.e., Landsat) was conducted to automatically extract disturbed forest area, including slope disaster such as debris flow and slope failure. In addition, time-series interferometric SAR (InSAR) using Phased-Array L-band Synthetic Aperture Radar-2 (PALSAR-2) visualized centimeter-scale displacement of the sloping area for the last decade. The displacement polygons, which are assumable to be susceptible area of future landslide, were delineated based on GIS analysis. Field survey in several accessible test sites showed that 50~60% of the extracted area truly indicated displacement signatures on the infrastructure. Detailed comparison of the spatial distribution of the susceptible polygons with geomorphological features in Sasebo region revealed that the importance of the geological background such as rock type, accumulation history of the past landslide debris, and inclination and direction of the bedding plane.